Gamma Spectrometry Forum

Hello there,
First, that's a pretty lame title for the thread, I know!
Today I report about another 28 day long measurement on the Trintite Type 5 sample from Mineral Research Company.
Apparently they ran out of Type5 sample now so I was lucky to get one of the last ones, as discussed in an earlier thread (viewtopic.php?f=5&t=755&sid=2dbb0780328 ... e4d39ee9ae) this specimen is unlike any other I tested before, not that it’s particularly hot, I tested hotter ones, but the spectrum is very rich. That’s allegedly because of the proximity of these “type 5” samples to the blast, which implies they were subject to a more intense neutron flux and therefore they are rich of activation products.

This is obvious looking at how high the Eu152 peaks are, but what really surprised me the first time around was the 81 keV peak, which I eventually labelled as Ba133, The bomb had an explosive lens system containing Baratol, which in turn contained stable Ba132. Activation due to neutron capture created Ba133, also visible in other Trinitite spectra.
There was some debate on that peak, due to its proximity to the peak expected from the lead shield fluorescence and also to the possibility of the presence of Pb212 and Pb214 in the sample.

So I decided to repeat the measurement improving my shielding with two pewter sheets of 15 x 75 cm, which allowed me to add roughly 4 mm more pewter to the shield. Pewter is pretty good at absorbing the XRF from the shielding. I didn’t expect the additional 4 mm of pewter to bring an improvement of the same magnitude as the first 4 mm of pewter did, but I expected to get better.
Here you can judge the result. The first picture is my shielded background from a recent measurement with 4 mm of pewter.

And this is the shielded background with 8 mm of pewter. I have to add I slightly improved the overall shielding which decreased from 61 to 59 CPS (my unshielded background in my test room is typically 255-260 CPS with this probe).
Now that doesn’t look like a dramatic improvement at first sight but you can see the XFR peak stands out less compared with the rest of the background spectrum in the picture below, so the effect of fluorescence is further reduced. The measurement started on June 3, lasted until July 1, and consisted in:
- 7 days of background.
- 14 days with the sample in the chamber.
- Further 7 days of background.
The 14 day background spectrum was later subtracted from the sample + background spectrum,

The result is the best I had so far, it’s smoother and clearer for one thing and the 81 keV peak is loud and clear. I really don’t think fluorescence from the shield can account for such a strong peak. I tested more active Trinitite specimens with the previous version of my shielding (4mm of pewter instead of 8 mm) and nothing like that ever showed up in the XRF region, not even close.
Spectrum is presented in counts per bin and energy per bin views.
I’ve recently been in touch with William Kolb, the author of the book “Trinitite: The Atomic Age Mineral”. You can reach him at syntec@verizon.net and, if interested, you can buy the book directly from him.
He’s also among the authors of this paper on Trinitite [broken link removed - Steven]
He had the following comments on the specimen and the spectrum:
“You were indeed lucky to get the high Eu152 specimens from MinResCo. The only specimens I’ve seen with this much Eu152 came from the shed at Trinity where the original surface was preserved. None of the usual eBay sellers has anything like it."
He added that another author of the paper (J.R) “has an excellent hpGe system. I am not sure what the 81 KeV peak is on my system because I don’t see other Ba133 peaks but he says he always sees the 81 KeV peak when there is Ba133 present and it definitely shows up on other hpGe systems.”
He also mentioned another measurement conducted by the other author of the paper (J.R), and I found this part particularly interesting:
“He mounted a large specimen of Trinitite and ground off one millimeter at a time. He measured the gamma activity in each fraction and plotted that. Fission products (Am241, Ba133, Cs137) all decreased rapidly from layer to layer while activation products and naturally occurring nuclides (Eu152,K40) stayed the same. We considered this pretty strong evidence that Ba133 comes mainly from Baratol and the low amount of Ba133 that I subsequently found in rock tends to confirm this.”

This was another reason for me to attribute that 81 keV mainly to Ba133.

I also want to show another potential piece of evidence that this is the case, in the following photo you see the spectrum both with and without background subtracted, the one in the upper half of the image includes background.
The vertical red line in both spectra corresponds to the 81 keV peak (actually 81.6 keV) in the spectrum with the background subtracted. When this line is transported in the spectrum where background is not subtracted it doesn't really match the XRF peak’s centroid, which seats around 77 keV as expected.

Putting all together I think Ba133 is by far the most likely candidate for that peak.

Again, there seems to be something else close to that peak around 85-86 leV but I decided to leave it unlabeled.

As for the rest, I identified a couple more Eu152 peaks, the one at 964 keV pretty clear, the other at 1213 keV much less obvious and I am not sure labelling it was the right thing to do. I mean, Eu152 is there, but if that peak is visible then the one at 1299 keV should be visible too and I see no sign of it.

The 1400-1500 keV peak drifted “north” in the final days of the acquisition due to thermal drifting. Last week weather got hotter here in Italy, as a consequence in the final days of accumulation of the background the K40 peak moved towards slightly lower energies and when subtracted that resulted in a higher energy difference peak (around 1475 keV). It’s not the first time this happens to me with Trinitite, the difference peak in that region is so weak it takes very little to make it “move” one way or another. It still comes from K40 (NORM) and Eu152 (activation).

One final point on the 13 keV peak. As you know the Trinity bomb was a Plutonium device so it wasn’t straightforward to me why I found an Uranium X-rays peak in it. In the end I thought Uranium could come from Plutonium decay so I went checking its nuclear data (http://nucleardata.nuclear.lu.se/toi/nu ... iZA=940239) which links that X-ray to Pu239, another indication of the presence of “unburned” plutonium in the specimen. Anyway, its contribution to the peak around 40 keV must be very small.

Finally, the quantitative analysis confirms the results of previous tests, the only difference is the slight improvement in terms of shielding which brought my shielded background to 59 CPS vis 255-260 CPS of my unshielded one.
This was the fourth measurement of this sample. I cumulatively spent 17+21+28+28 = 94 days of accumulation time since January on it. Now I can finally say I am happy with it….for the time being at least!

Have a nice weekend.

Massimo

Massimo,

This is a beautiful job of research and presentation. You have convinced me that the peak at 81 keV is Ba133. I think your specimen is pretty special. I have three Trinitite samples and while the best one resembles this one, none of them have anything like that peak at 81.

The predominate peak for lead florescence is the K Alpha 1 characteristic x-ray at 75 keV. Your peak at 81 is in the wrong place and has the wrong shape for lead florescence.

Massimo,

I'm a fan of creative thread titles, so I like yours.

That is such an interesting and well-prepared spectrum. You clearly put a lot of care and effort into your work. It will be a nice example for me as I continue to learn. I have a piece (from a friend, ID: JC-TR-01 in some of my threads) that has a similar Eu peaks and there is a small bump at about 80 keV that I'll now work harder to confirm.

I am curious about the lead peak in your background. You should have many times the amount of pewter needed to wipe out the lead x-ray. Gilmore states 3mm of tin to reduce the lead x-rays by 1000-fold. The pewter I used has 92% tin, 7.5% antimony, and 0.5% copper; I have 3mm of this pewter and 2mm of copper and there's no lead peak that I've been able to see. I looked up that antimony has an x-ray similar to tin (but I'm not sure). Is it possible your pewter is a tin-lead type? What do you use for end(s) of your shield (does it have pewter and copper)? Sorry for all the questions; I'm completely re-doing my shield, so I'm trying to learn and understand as much as possible before I buy more materials.

The paper on Trinitite you linked is great, it's the best low energy spectrum of Trinitite I've seen and very helpful. I've added it to my "library". I also have the Trinitite book by Kolb, it's well worth the cost.

Again, great job, and thank you for sharing.

Mike S.

Massimo,

Nice work and thanks for sharing.

Just a note on shielding that isn't often discussed. Increasing the internal diameter of your shield also decreases the counts from fluorescence, this is why you see professional shields from the likes of Canberra look like big buckets.

With a larger internal diameter some of the reflected energy simply misses the detector and get's absorbed in the walls of the chamber.

The obvious problem is cost of lead and the inconvenience of having something that heavy.

Steven

We are playing with low resolution detector, even with good energy calibration, under 100keV are nowhere linear and resolution is quite horrible as a well known behaviour stated many times before.

Even in Pittauerova works with HPGe Ba133 lines they just barely emerges from Pb Ka/Kb x-rays lines in a professional rig...

If Ba133 lines are here they are well buried in Pb fluorescence, also the Ba133 sum peak is missing, imho normal Pb lines are shown.

cicastol wrote: ↑

10 Jul 2020, 06:14

We are playing with low resolution detector, even with good energy calibration, under 100keV are nowhere linear and resolution is quite horrible as a well known behaviour stated many times before.

Even in Pittauerova works with HPGe Ba133 lines they just barely emerges from Pb Ka/Kb x-rays lines in a professional rig...

If Ba133 lines are here they are well buried in Pb fluorescence, also the Ba133 sum peak is missing, imho normal Pb lines are shown.

You bring up a good point that I hadn't considered: Sum peaks can be used to assist in identifying or confirming the source of a peak.

I'm looking at the Pittauerova spectrum and I don't see any Ba133 sum peaks that would be detectable in Massimo's spectrum. Obviously they used a different sample, but he's getting pretty good alignment with them if you assume his threshold of detection is roughly equivalent to 0.1 imp/s on the Pittauerova spectrum.

To be clear: I'm just trying to understand, I'm new to gamma spectrometry.

cicastol wrote: ↑

10 Jul 2020, 06:14

We are playing with low resolution detector, even with good energy calibration, under 100keV are nowhere linear and resolution is quite horrible as a well known behaviour stated many times before.

Even in Pittauerova works with HPGe Ba133 lines they just barely emerges from Pb Ka/Kb x-rays lines in a professional rig...

If Ba133 lines are here they are well buried in Pb fluorescence, also the Ba133 sum peak is missing, imho normal Pb lines are shown.

Sorry for not answering all comments at once, I’ll do that later, now I focus on the Ba133 issue:

My point is that, after testing seven Trinitite samples of different activity over nearly 12 months, I don’t see how that peak could be from XRF alone, because nothing like that ever emerged from samples which should have given me even higher fluorescence peaks, I am going to show results from two tests to highlight that:

First test
A sample of three very small specimens which I tested together. Back when I did this test I didn’t have any pewter in my lead shield and you can see from the background spectrum how much higher the fluorescence peak was compared with what I get now with my current shielding, so with this setup I was much more prone to get a XRF peak.
The sample gave 8.90 CPS in average, therefore quite close to what the type 5 specimen gave in my last measurement.
In summary there were almost as many gammas hitting my shielding as in my latest test and nothing to prevent the resulting XRF from being detected by my probe and enter my spectrum.
Also, as Steven righly pointed out, the absence of pewter also meant that the lead was physically closer to the scintillator.

Still the result didn't show anything clearly emerging from the low energy slope. And yes, that was a shorter measurements, but when there’s something there it emerges well before 24 hours, the extra accumulation time is just in order to make the result look smoother.
Second test
This time I had 4 mm of pewter which reduced XRF considerably (see first post of the thread), but still not as much as in my last test (where I had 8 mm of it) and the specimen was quite more active than the type 5 I tested in my last measurement, 15.55 CPS in average vs 10.40, so basically 50% more counts, with a considerably stronger Cs137 peak, and a very similar background (61 CPS vs 59 CPS).
So, a lot more counts, most from Cs137, and a less efficient shield to block XRF from entering the spectrum.
Still, in the result all you can see is the hint of a small bump, nothing like what I got from the other sample in 80 keV region.
My previous attempts with other samples left me quite unsatisfied with the result, but I am happy I did them because they taught me something about what to expect.

There’s no doubt that some XRF is there, just like some Pb212 and Pb214, all of them giving counts in their usual regions, but this is the case for all Trinitite samples, and in my experience those contributions, although present, are not enough to explain such a loud and clear peak. This is not consistent with what I learned in the Trinitite measurements I did in the last 12 months. In my experience there must be something else to make the spectrum look like that.
I wasn’t looking for Ba133 when I first tested this sample, but I couldn’t ignore what I saw, particularly when others, with a better equipment and more experience in the field (both in the field of gamma spectroscopy and in that of Trinitite) saw the same.
And how much Ba133 contributes to that peak really depends on the sample, on that respect every sample is different. Before coming across this sample I didn't even consider the possibility to detect it.

Am I 100% sure? of course not, you always have to doubt everything, particularly your own results (in the words of Richard Feynman “The first principle is that you must not fool yourself, and you are the easiest person to fool”) but I think I have enough elements to say Ba133 is the most likely candidate.
I am ok with you having a different opinion on the matter, actually I welcome that, it’s the kind of things that push you to look deeper, which is what I tried to do with this measurement and who knows, maybe this is not the end of it.
I doubt I’ll ever find a better Trinitite sample, so I’ll probably test it again and again, as I improve my shielding.

Massimo

I just did a quick run on my Trinitite specimen. Here are plots from PRA and Thermino for reference. Looks pretty consistent with the others. There are a lot of fake "Trinitite" samples out there since it is illegal to collect nowadays but the presence of Eu-152 and Am-241 peaks along with Cs-137 are evidence that my sample occurred under a mushroom cloud.





Best,
Andrey

Mike S wrote: ↑

06 Jul 2020, 08:09

I am curious about the lead peak in your background. You should have many times the amount of pewter needed to wipe out the lead x-ray. Gilmore states 3mm of tin to reduce the lead x-rays by 1000-fold. The pewter I used has 92% tin, 7.5% antimony, and 0.5% copper; I have 3mm of this pewter and 2mm of copper and there's no lead peak that I've been able to see. I looked up that antimony has an x-ray similar to tin (but I'm not sure). Is it possible your pewter is a tin-lead type?

Yeah, that might well be the case, actually when I bought my pewter i didn't investigate that detail, which I certainly will.

What do you use for end(s) of your shield (does it have pewter and copper)? Sorry for all the questions; I'm completely re-doing my shield, so I'm trying to learn and understand as much as possible before I buy more materials

My shield is far from finished, right now the end of the room only have a few mm of lead, it's currently the weakest part of my shielding.

I think your specimen is pretty special.

Yeah, I was lucky to get one of the very last Type5 samples available, I don't know if they were all like that. There are no more available so I won't be able to answer that question.